Anda di halaman 1dari 5

IJSTE - International Journal of Science Technology & Engineering | Volume 3 | Issue 09 | March 2017

ISSN (online): 2349-784X

Automatic Traffic Rule Violation Detection and


Number Plate Recognition
Amey Narkhede Vikrant Nikam
Student Student
Sinhgad College of Engineering, Pune, India Sinhgad College of Engineering, Pune, India

Akshay Soni Abhishek Sathe


Student Student
Sinhgad College of Engineering, Pune, India Sinhgad College of Engineering, Pune, India

Prof. G.G.Chiddarwar
Assistant Professor
Sinhgad College of Engineering, Pune, India

Abstract
In the new evolving world, trafc rule violations have become a central issue for majority of the developing countries. The numbers
of vehicles are increasing rapidly as well as the numbers of traffic rule violations are increasing exponentially. Managing traffic
rule violations has always been a tedious and compromising task. Even though the process of trafc management has become
automated, its a very challenging problem, due to the diversity of plate formats, different scales, rotations and non-uniform
illumination conditions during image acquisition. The principal objective of this project is to control the traffic rule violations
accurately and cost effectively. The proposed model includes an automated system which uses IR sensors and camera based on
Arduino to capture video. The project presents Automatic Number Plate Recognition (ANPR) techniques and other image
manipulation techniques for plate localization and character recognition which makes it faster and easier to identify the number
plates. After recognizing the vehicle number from number plate the SMS based module is used to notify the vehicle owners about
their traffic rule violation. An additional SMS is sent to Regional Transport Office (RTO) for tracking the report status.
Keywords: ANPR, IR Sensor, SMS, RTO
________________________________________________________________________________________________________

I. INTRODUCTION

Automation in day to day life has gained importance in recent years. The number of accidents on the roads is due to the rule
violations such as breaking traffic signals, over-speeding, driving on wrong sides etc. To avoid such traffic violations, traffic police
has to be present on the road and has to continuously check if some vehicle is violating the rule. A certain automated solutions
were developed to eliminate the violations; however each of them had certain limitations. For example, the video capturing cameras
eliminated need of a authority to be present to check rule violation. However, whole stored video had to be checked manually for
the rule violation scenario.[4] In this proposed system, a solution for signal breaking violation is given .The system includes an
automated system by using IR sensor, camera and number plate recognition application.[3] In this system IR sensor will be placed
near zebra crossing line. If any vehicle crosses the zebra line, the desktop application will be initiated and will capture number
plate image. Number plate recognition application by using image processing algorithm will recognize number plate and SMS will
be sent to the offender in case of rule violation scenario.[2]

II. GUIDELINES FOR MANUSCRIPT PREPARATION

Abbreviations and Acronyms


Arduino
Arduino is an open-source platform used for building electronics projects. Arduino consists of both a physical programmable
circuit board (often referred to as a microcontroller) and a piece of software, or IDE (Integrated Development Environment) that
runs on your computer, used to write and upload computer code to the physical board. The Arduino platform has become quite
popular with people just starting out with electronics, and for good reason.
Bluetooth
Bluetooth is a telecommunications industry specification that describes how mobile phones, computers, and personal digital
assistants can be easily interconnected using a short-range wireless connection. Using this technology, users of phones, pagers,
and personal digital assistants can buy a three-in-one phone that can double as a portable phone at home or in the office, get quickly

All rights reserved by www.ijste.org 559


Automatic Traffic Rule Violation Detection and Number Plate Recognition
(IJSTE/ Volume 3 / Issue 09 / 112)

synchronized with information in a desktop or notebook computer, initiate the sending or receiving of a fax, initiate a print-out,
and, in general, have all mobile and fixed computer devices be totally coordinated.

III. MATH

Three main algorithms are being used, first algorithm is edge detection in an image is to find object boundaries, which in turn can
reduce the amount of data to be analyzed in an image, second algorithm the Hough transform is a feature extraction technique used
in image analysis, computer vision, and digital image processing.. The last algorithm is used to recognize the segmented characters
which is performed on extracted license plate image. The three algorithms are mentioned namely:

Edge Detection
Although certain literature has considered the detection of ideal step edges, the edges obtained from natural images are usually
not at all ideal step edges. Instead they are normally affected by one or several of the following effects:
Focal blur caused by a finite depth-of-field and finite point spread function.
Penumbral blur caused by shadows created by light sources of non-zero radius.
Shading at a smooth object
A number of researchers have used a Gaussian smoothed step edge (an error function) as the simplest extension of the ideal
step edge model for modeling the effects of edge blur in practical applications[4].
We can use a periodical convolution of the function f with specific types of matrices m to detect various types of edges in an
image:
f (x,y ) =f(x,y )* m[x,y ]= 1 1
=0 =0 f(x,y)* m[mod(x-i),mod(y-j)]
Where w and h are dimensions of the image represented by the function f.

Fig. 1: Edge Detection [2]

Hough Transform
It is a feature extraction technique initially used for line detection. Later on it has been extended to find position of arbitrary shape
like circle or oval. The linear Hough transform algorithm uses a two-dimensional array, called an accumulator, to detect the
existence of a line described by r=xcos + . The dimension of the accumulator equals the number of unknown parameters,
i.e., two, considering quantized values of r and in the pair (r,). For each pixel at (x,y) and its neighborhood, the Hough transform
algorithm determines if there is enough evidence of a straight line at that pixel. If so, it will calculate the parameters (r,) of that
line, and then look for the accumulator's bin that the parameters fall into, and increment the value of that bin. By finding the bins
with the highest values, typically by looking for local maxima in the accumulator space, the most likely lines can be extracted, and
their (approximate) geometric definitions read off. The simplest way of finding these peaks is by applying some form of threshold,
but other techniques may yield better results in different circumstances determining which lines are found as well as how many.
Since the lines returned do not contain any length information, it is often necessary, in the next step, to find which parts of the
image match up with which lines. Moreover, due to imperfection errors in the edge detection step, there will usually be errors in
the accumulator space, which may make it non-trivial to find the appropriate peaks, and thus the appropriate lines. The final result
of the linear. Hough transform is a two-dimensional array (matrix) similar to the accumulatorone dimension of this matrix is
the quantized angle and the other dimension is the quantized distance r. Each element of the matrix has a value equal to the sum
of the points or pixels that are positioned on the line represented by quantized parameters (r, ). So the element with the highest
value indicates the straight line that is most represented in the input image.

All rights reserved by www.ijste.org 560


Automatic Traffic Rule Violation Detection and Number Plate Recognition
(IJSTE/ Volume 3 / Issue 09 / 112)

Fig. 2: Hough Transform [9]

K-Nearest Neighbour
In pattern recognition, the k-nearest neighbors formula (k-NN) may be a non-parametric methodology used for classification and
regression.[1] In each cases, the input consists of the k nearest coaching examples within the feature area. The output depends on
whether or not k-NN is employed for classification or regression: In k-NN classification, the output may be a category membership.
AN object is classed by a majority vote of its neighbors, with the thing being assigned to the category most typical among its k
nearest neighbors (k may be a positive number, usually small). If k = 1, then the thing is just assigned to the category of that single
nearest neighbor. In k-NN regression, the output is that the property worth for the thing. This worth is that the average of the values
of its k nearest neighbors-NN may be a style of instance-based learning, or lazy learning, wherever the operate is simply
approximated domestically and every one computation is delayed till classification. The k-NN formula is among the best of all
machine learning algorithms. Both for classification and regression, it may be helpful to assign weight to the contributions of the
neighbors, in order that the nearer neighbors contribute additional to the typical than the additional distant ones. As an example, a
standard coefficient theme consists in giving every neighbor a weight of 1/d, wherever d is that the distance to the neighbor. The
neighbors are taken from a collection of objects that the category (for k-NN classification) or the thing property worth (for k-NN
regression) is thought. This will be thought of because the coaching set for the formula, although no express coaching step is
needed. A defect of the k-NN formula is that it's sensitive to the native structure of the information. The formula isn't to be confused
with k-means, another standard machine learning technique.

Fig. 3: K Nearest Neighbour [6]

Architecture Flow

Fig. 4: Architectural Flow Diagram

All rights reserved by www.ijste.org 561


Automatic Traffic Rule Violation Detection and Number Plate Recognition
(IJSTE/ Volume 3 / Issue 09 / 112)

Desktop Application
1) IR sensor will sense if vehicle is present on zebra line when signal is red.
2) Camera will start after IR sensor senses vehicle. SCOE, Dept. of Computer Engineering 8 Year 2016-17
3) Camera will capture the footage of the vehicle.
4) Desktop application will be initiated and will capture the image.
5) 5.Localization of the number plate will initerte
6) Number extracted from the number plate will be associated with owners mobile number.
7) Rule violation message will be sent to the owner of the vehicle.

Constraints
1) Accuracy of the system might vary depending upon light intensity changes and distance between camera and object and format
(font) of number plate image
2) Effectiveness of cameras is inconclusive.

IV. REQUIREMENT SPECIFICATION

Software Requirements
Language: Java J2SE and JDK
IDE: NetBeans 8, Eclipse
Operating System: Windows 7
Database required: SQL
OpenCV library
Arduino Software Environment.

Hardware Requirements
IR Sensor
Arduino board
4 GB RAM
500 GB Hard disk
Breadboard
Jumper Wires
100k Resistor
Bluetooth serial Board
Camera-2 Megapixel I.P. PTZ camera with integrated I.R

V. CONCLUSION

In this review paper, the automatic number plate recognition system using vehicle license plate is presented. The system use image
processing techniques for identifying the vehicle from the database stored in the computer. The system works satisfactorily for
wide variation of conditions and different types of number plates. It is implemented and executed in OpenCV and performance is
tested on genuine images.
The system works quite well however, there is still room for improvement. The camera used in the system for this project is
sensitive to vibration and fast changing targets due to the long shutter time. The system speed can be increase with high resolution
camera. The character recognition method is sensitive to misalignment and to different sizes, so the affine transformation can be
used to improve the character recognition from different size and angles. The statistical analysis can also be used to define the
probability of detection and recognition of the vehicle number plate. At present there are certain limits on parameters like speed of
the vehicle, script on the vehicle number plate, skew in the image which can be removed by enhancing the algorithms further.

REFERENCES
[1] Chi-Hung Chuang ,Luo-Wei Tsai ,Ming-Shan Deng ,Jun-Wei Hsieh ,Kuo-Chin Fan, Vehicle Licence Plate Recognition Using Super-Resolution Technique
, 2014, IEEE
[2] Khalil M. Ahmad Yousef, Maha Al-Tabanjah, Lia Morra and Silivia Chiusano, Automatic Number Plate Recognition System using Super Resolution
Technique , 2015, IEEE
[3] Amiya Kumar Tripathy, Rebeck Carvalho, Keshav Pawaskar, Suraj Yadav, Vijay Yadav, Automatic Number Plate Recognition System for vehicle
identification using Optical Character Recognition , 2015, IEEE
[4] http://www.onlinejournal.in/IJIRV2I5/050.pdf, Violation Detection at Traffic Signals Using RFID System
[5] http://airccse.org/journal/ijcsea/papers/3313ijcsea04.pdf, Design and development of gps-gsm based tracking system with google map based monitoring

All rights reserved by www.ijste.org 562


Automatic Traffic Rule Violation Detection and Number Plate Recognition
(IJSTE/ Volume 3 / Issue 09 / 112)

[6] Manish Kumar,M K Jindal, K-Nearest Neighbour based offline handwritten Gurmukhi Character Recognition , 2011, International Conference on Image
Processing
[7] Ascar Davis X,Oshin V J,Laura,Shamali P, License plate localization by sobel vertical edge detection method , 2012, International journal of Electrical
Engineering and Research
[8] Balamurugan G,Rakishly K, Automatic Number plate Recognition System using super Resolution Technique , 2014 ,IEEE
[9] Ondrej Martinsky,Algorithmic and mathematical principles of automatic number plate recognition systems,2010 ,BRNO University
[10] https://en.wikipedia.org.

All rights reserved by www.ijste.org 563